Efficient R-estimation of principal and common principal components

Simple item page
dc.contributor.authorHallin, Marc
dc.contributor.authorPaindaveine, Davy
dc.contributor.authorVerdebout, Thomas
dc.date.accessioned2013-04-08T13:35:39Z
dc.date.available2013-04-08T13:35:39Z
dc.date.issued2013-04-08
dc.description.abstractWe propose rank-based estimators of principal components, both in the one- sample and, under the assumption of common principal components, in the m- sample cases. Those estimators are obtained via a rank-based version of Le Cam's one-step method, combined with an estimation of cross-information quantities. Under arbitrary elliptical distributions with, in the m-sample case, possibly heterogeneous radial densities, those R-estimators remain root-n consistent and asymptotically normal, while achieving asymptotic e ciency under correctly speci ed densities. Contrary to their traditional counterparts computed from empirical covariances, they do not require any moment conditions. When based on Gaussian score functions, in the one-sample case, they moreover uniformly dominate their classical competitors in the Pitman sense. Their finite-sample performances are investigated via a Monte-Carlo study.en
dc.identifier.urihttp://hdl.handle.net/2003/30135
dc.identifier.urihttp://dx.doi.org/10.17877/DE290R-5362
dc.language.isoende
dc.relation.ispartofseriesDiscussion Paper / SFB 823;12/2013
dc.subjectcommon principal componentsen
dc.subjectelliptical densitiesen
dc.subjectprincipal componentsen
dc.subjectranksen
dc.subjectR-estimationen
dc.subjectrobustnessen
dc.subjectuniform local asymptotic normalityen
dc.subject.ddc310
dc.subject.ddc330
dc.subject.ddc620
dc.titleEfficient R-estimation of principal and common principal componentsen
dc.typeTextde
dc.type.publicationtypeworkingPaperde
dcterms.accessRightsopen access

Files

Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
DP_1213_SFB823_Hallin_Paindaveine_Verdebout.pdf
Size:
1.28 MB
Format:
Adobe Portable Document Format
Description:
DNB
Download
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
1.02 KB
Format:
Item-specific license agreed upon to submission
Description:
Download

Collections

Sonderforschungsbereich (SFB) 823